Transforming Growth Factor-/?
(TGF/?) Inhibits TGFa Expression in
Bovine Anterior Pituitary-Derived
Cells
Susan G. Mueller and Jeffrey E. Kudlow
Department of Clinical Biochemistry
University of Toronto (S.G.M.)
Toronto, Ontario, Canada
Department of Medicine
Division of Endocrinology
University of Alabama (J.E.K.)
Birmingham, Alabama 35294
Transforming growth factor-/91 (TGF/31) is a multifunctional regulator of cell growth and differentiation. We report here that TGF01 decreased the proliferation of nontransformed bovine anterior pituitary-derived cells grown in culture. We have
previously demonstrated that these cells express
both TGFa and its receptor [the epidermal growth
factor (EGF) receptor] and that expression can be
stimulated by phorbol ester (TPA) and EGF. TGF01
treatment over a 2-day period decreased the proliferation of pituitary cells. This decreased growth rate
was accompanied by a decrease in the TGFa mRNA
level. The effect of TGF/S1 on TGFa mRNA downregulation was both dose dependent (maximal effect
observed at 1.0 ng/ml TGF01) and time dependent
(minimum of 2-day treatment with TGF/91 was required before a decrease in TGFa mRNA was observed). Studies on TGFa mRNA stability indicated
that TGF01 did not alter the TGFa mRNA half-life.
Treatment of the TGF01 down-regulated cells with
EGF resulted in the stimulation of TGFa mRNA levels; thus, the TGF01 -treated cells remained responsive to EGF. The decreased proliferation in response
to TGF/31 could be only partially reversed by simultaneous treatment of the cells with EGF (10~9 M) and
TGF/S1 (3.0 ng/ml). Qualitatively, the TGF/31-induced
reduction of TGFa mRNA content was independent
of cell density. TGF01 treatment of the anterior pituitary-derived cells also reduced the levels of cmyc and EGF receptor mRNA. These results represent the first demonstration of the down-regulation
of TGFa synthesis by a polypeptide growth factor
and suggest that TGF01 may be a physiological
regulator of TGFa production in vivo. (Molecular Endocrinology 5: 1439-1446, 1991)
INTRODUCTION
Transforming growth factor-a (TGFa), a member of the
epidermal growth factor (EGF) family of mitogenic polypeptides, is generally capable of stimulating the proliferation of both epithelial and mesenchymal cells (1).
Although TGFa shares only a 32% amino acid sequence
homology with EGF (2), both bind to and elicit their
biological effects through the same receptor, the EGF
receptor (3-8). Because TGF« was initially detected in
the conditioned medium of transformed cells (4) and
later in fetal and embryonic tissue (9, 10, 11), it was
believed to be an onco-fetal form of EGF (12). More
recently, TGFa has been identified in a variety of normal
adult tissues, including the pituitary (13), brain (14-16),
ovary (17), keratinocytes (18), macrophages (19, 20),
and vascular smooth muscle cells (21). The presence
of TGFa in these tissues implies a role for TGFa in
normal physiological processes. Since the growth of
normal cells is under tight regulation, the expression of
TGFa in normal tissues must also be regulated.
To study the regulation of TGFa in normal cells, we
used cells cultured from bovine anterior pituitary glands.
We have previously shown that TGFa is localized in the
somatotrophs and lactotrophs within the anterior pituitary gland (13), and cells derived from these glands
secrete TGFa into their culture medium (8, 22). The
expression of TGFa can be up-regulated in these cells
by exposure to EGF and the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) (23). The ability of
EGF to stimulate TGFa synthesis suggests that TGFa
can stimulate its own synthesis by an autocrine or a
paracrine route. TGFa has been shown to autostimulate
its own production in other cell systems, including the
human breast cancer cell line MDA 468 (24) and nontransformed primary keratinocytes (18).
0888-8809/91/1439-1446$03.00/0
Molecular Endocrinology
Copyright © 1991 by The Endocrine Society
The ability of a growth factor to stimulate its own
production could, if left unchecked, result in over1439
MOL ENDO-1991
1440
Vol5No. 10
expression of that growth factor. Recent studies have
demonstrated that TGFa overexpression, driven by a
strong heterologous promoter in either transgenic mice
(25-27) or transfected Rat-1 fibroblasts injected into
nude mice (28) resulted in pathological hyperplasia or
tumor formation in the respective models. The ability of
normal tissues to control TGFa expression, despite the
existence of an autostimulatory mechanism, necessitates the coexistence of a negative regulatory pathway
capable of down-regulating growth factor expression.
The present study was undertaken to determine
whether TGF/31, a potent epithelial cell growth inhibitor
(29, 30), was capable of negatively regulating TGFa
expression. We report here that TGF/31 inhibited both
TGFa and EGF receptor expression and decreased the
growth rate of the anterior pituitary cells in culture. This
is the first demonstration of the down-regulation of
TGFa production by a polypeptide growth factor, and
it suggests TGF/31 may be a physiological regulator of
TGFa production in normal tissues.
RESULTS
We investigated the effect of TGF01 on the growth rate
of bovine anterior pituitary-derived cells grown in culture. Over the first 2 days, the cells treated with either
1 or 3 ng/ml TGF/31 grew at the same rate as untreated
cells, but subsequently, the TGF/31-treated cells became essentially growth arrested (Fig. 1), while the
untreated cells continued to proliferate.
Northern blot analysis was carried out to determine
whether TGF/31 could modulate the content of TGFa:
400
•
300
NoTGFB
X 1.0 ng/ml TGFG
A 3.0 ng/ml TGF8
mRNA in the pituitary cells. Cells were subcultured at
a low density and treated with increasing doses of
TGF/31 (0-3.0 ng/ml) for 4 days. Culture medium, in the
absence or presence of TGF/31, was replaced daily.
Northern blot analysis of poly(A)+ RNA indicated that
TGF/31 treatment resulted in a decreased content of
TGFa mRNA in these cells (Fig. 2). This decrease in
TGFa mRNA was dose dependent, with a maximal
decrease observed at a concentration of 1.0 ng/ml
TGF/31. Densimetric analysis indicated that the decrease in TGFa mRNA resulting from treatment with 1
ng/ml TGF/31 was about 70%. The level of actin mRNA
was unchanged during the course of this treatment
(Fig. 2). The effect of TGF/31 on the pituitary cells was
also time dependent. Cells were treated daily with 1.0
ng/ml TGF/31 for the indicated periods of time. The
initiation of TGF/31 treatment was staggered, such that
all treatments ended simultaneously (Fig. 3). Northern
blot analysis indicated that a decrease in the level of
TGFa mRNA was detected only after a 2-day treatment
of the cells with TGF01 (Fig. 3). Reprobing of this
Northern blot with c-myc revealed that TGF/31 treatment also resulted in a decrease in the c-myc mRNA,
but this decrease was first apparent after 1 day of
treatment (Fig. 3). The decrease in TGFa and c-myc
mRNA was shown to be relatively specific, as the level
of hexosaminidase-A mRNA was not altered during the
TGF/31 treatment.
Other studies have indicated that the effect of TGF/31
on smooth muscle cell proliferation is density dependent
(31). To determine whether cell density had an effect
on the response of the pituitary cells to TGF/31, cells
were subcultured at densities to approximate 10% and
70% confluence. The sparse cells were treated with
TGF/31 (1.0 ng/ml) for 4 days, while the more dense
cells were treated for 3 days. Culture medium in the
absence or presence of TGF/31 was changed daily.
TGF/31 treatment of these cells decreased the content
of TGFa mRNA at both cell densities after treatment
for the indicated lengths of time (Fig. 4), although the
effect was more pronounced in the sparse cells.
0 0.1 05 1.0 3.0
200
ng/ml TGF-p
TGF-a
Actin
Time (Days)
Fig. 1. Effect of TGF/31 on Proliferation of Cultured Bovine
Anterior Pituitary-Derived Cells
The pituitary cells were subcultured at a density of 2.5 x
104 cells/35-mm culture dish in 2% calf serum-DMEM. After
an overnight incubation, the medium was replaced with serumfree medium alone or with 1.0 or 3.0 ng/ml TGF/31. Cells
cultured in the presence of TGF/31 were given fresh growth
factor daily. Cells were counted on days 0-4, using a Coulter
counter. Each point represents the mean result of triplicate
determinations (average SD, <5%).
Fig. 2. Effects of Varying Doses of TGF/31 on the TGFa mRNA
Level in Bovine Anterior Pituitary-Derived Cell Cultures
Sparse pituitary cells were treated for 4 days with varying
doses of TGF/31 (0-3.0 ng/ml). The medium was replaced
daily in the absence or presence of TGF01. After the 4-day
treatment, the RNA was extracted. The polyadenylated RNA
(2 Mg/lane) was electrophoresed in a 1 % agarose-6% formaldehyde gel. The RNA was transblotted onto a nylon membrane
and probed simultaneously with the human TGFa cDNA and
the chicken actin cDNA.
1441
TGF/31 Down-Regulates TGFa
0 .25
TGF
I
2
-* - mmm •
4 d
•
Hex A _
Fig. 3. Effect of Duration of TGF01 Treatment on TGFa and
c-myc mRNA Levels in Cultured Bovine Anterior PituitaryDerived Cells
Bovine anterior pituitary-derived cells were subcultured at a
low density onto collagen-coated tissue culture plates. Cells
were given fresh medium with or without TGF/31 (1.0 ng/ml)
daily. Cells receiving TGF/31 for 4 days were given fresh
medium with TGF/31 daily. The introduction of TGF/31 to the
cultures was staggered, such that all incubations, regardless
of duration, ended at the same time. Two micrograms of
polyadenylated RNA were analyzed by Northern blot, and the
nylon membrane was probed simultaneously with human
TGFa and hexosaminidase-A cDNAs. After autoradiography,
the hybridized cDNA probes were eluted from the Northern
blot, and the blot was reprobed with the c-myc cDNA.
would be observed with sparse secondary cultures.
The effect of TGF01 under these conditions was also
examined. Sparse cell cultures were either allowed to
condition their medium for 4 days, or the medium was
changed daily. These cells were also cultured in the
presence or absence of TGF/31 (1.0 ng/ml) throughout
the 4-day period. Northern blot analysis indicated that,
unlike the confluent secondary cell cultures, the sparse
cells, which were allowed to remain in their conditioned
medium, did not contain elevated levels of TGFa mRNA
(Fig. 5). TGF/31 treatment under both conditions resulted in a decreased accumulation of TGFa mRNA.
TGF/31 treatment also resulted in a decrease in both
the EGF receptor and c-myc mRNA levels. These
changes were specific, as the level of hexosaminidaseA mRNA was unaffected by the various treatments (Fig.
5). Analysis of the 4-day conditioned medium by the
EGF RRA indicated that cells cultured in the absence
of TGF/31 had approximately 10~10 M EGF-displacing
activity, while the cells treated with TGF/31 had less
than 10"11 M EGF-displacing activity in the medium (data
not shown).
We have established that EGF can stimulate the
accumulation of TGFa mRNA in pituitary cells (23) and
CD
CM
TGF-p
EGFR
TGF-oc
TGF-a
Hex A
m «*
Fig. 4. Effect of Cell Density on the Response to TGF/31 of
Cultured Bovine Anterior Pituitary-Derived Cells
Anterior pituitary derived-cells were subcultured at densities
to approximate 1) 10% confluence and 2) 70% confluence.
Cells plated at the lower density (1) were treated with (+) or
without (-) TGFj81 for 4 days. The more dense cells (2) were
treated in the absence (-) or presence (+) of TGF/31 for 3
days. Fresh medium was added to the cells daily. Two micrograms of polyadenylated RNA extracted from these cells were
analyzed by Northern blot, and the nylon membrane was
probed simultaneously with the human TGFa and the hexosaminidase-A cDNAs.
In earlier studies w e have shown that confluent cultures of the pituitary cells, which had been allowed to
condition their medium for 3 days, contained elevated
levels of TGFa mRNA relative to cell cultures whose
medium was changed daily (23). A similar experiment
was conducted to determine whether the same result
Myc
Hex A
Fig. 5. Effect of Allowing the Cells to Condition Their Medium
on the TGF/31 Responses
Bovine anterior pituitary-derived cells were subcultured at a
low density. Some cells (CD) received fresh medium with or
without TGF/31 (1.0 ng/ml) daily for 4 days. Other cells (CM)
were allowed to remain in their conditioned medium for 4 days.
TGF/31 was added daily to the culture medium of the cells on
conditioned medium, or they were left untreated for 4 days.
After the 4-day treatment, RNA was extracted from the cells.
Two micrograms of polyadenylated RNA were analyzed by
Northern blotting, and the nylon membrane was probed simultaneously with the human TGFa cDNA and the hexosaminidase-A cDNA. After autoradiography, the hybridized
cDNA probes were eluted from the Northern blot. The blot
was then reprobed with the EGF receptor (EGFR) cDNA and
the c-myc cDNA.
MOL ENDO-1991
1442
Vol5No. 10
MDA468 breast cancer cells (24). TGF/31 has been
shown to block some of the effects elicited by EGF in
other systems (32). We, therefore, investigated the
effect of TGF/31 on the stimulation of TGFa mRNA
accumulation by EGF. Cells were treated for 4 days in
the absence or presence of 1.0 ng/ml TGF/31. TGF/31 treated cells were then stimulated with 10~9 M EGF for
8 h. Stimulation of the TGF/31-treated cells with EGF
resulted in an increase in TGFa mRNA (Fig. 6). This
result indicates that TGF/31 treatment of cells does not
permanently impair their ability to express TGFa in an
EGF-responsive manner. We also determined whether
the growth inhibitory effect of TGF/31 could be reversed
by simulataneous treatment of the cells with EGF.
Figure 7 shows that in contrast to the effect on TGFa
mRNA (Fig. 6), EGF could not completely reverse the
growth inhibitory effect of TGF01, although it caused a
significant (P < 0.05), but small, increase in cell number
compared to that of cells treated with TGF/31 alone.
The TGFa mRNA is very stable in MDA468 cells (24),
and preliminary experiments had shown that it was also
stable in pituitary cells. A decrease in this stability could
result in a marked decrease in the cellular content of
TGFa, assuming a constant transcriptional rate. We,
therefore, investigated the effect of TGF/31 treatment
on the stability of TGFa mRNA in pituitary cells. Cells
were subcultured at a low density and allowed to grow
for 4 days. Cells treated with TGF/31 received 1.0 ng/
ml TGF/31 for the last 36 h of the 4-day incubation. This
duration of treatment was chosen because the level of
TGFa mRNA is decreased by 48 h. After this treatment
1 2
100
A No Growth Factors
80
•
TGFG
0 TGFB+EGF
-
60
Time (Days)
Fig. 7. Effect of EGF on the Growth of TGF/31-treated Cells
Pituitary cells were plated in replicate at low density and
then treated with or without TGF/31 (3.0 ng/ml), as described
in Fig. 1. Half of the TGF/31 -treated cells were treated with
EGF (10~9 M) for the duration of the experiment. Cell counts
were obtained daily for 7 days from triplicate cultures (average
SD,
- TGF-p
0
T6F- a _
Hex A
_!
Myc
_
3
5
+ TGF-p
8h
0
I
3
5
8h
•••• mm*
••in
3
TGF-a
Hex A
Fig. 6. Effect of EGF on the Down-Regulation of TGFa mRNA
by TGF01
The pituitary cells were subcultured at a low density and
given fresh medium daily with or without TGF/31 (1.0 ng/ml)
for 4 days. After the 4-day TGF/31 treatment, cells were
stimulated with 10~9 M EGF for 8 h in the presence of TGF/31
before RNA extraction. Two micrograms of polyadenylated
RNA were analyzed by Northern blotting, and the nylon membrane was probed simultaneously with human TGFa cDNA
and hexosaminidase-A cDNA. Lane 1, Untreated pituitary cells;
lane 2, TGF/31-treated cells; lane 3, TGF/31 -treated cells, followed by an 8-h stimulation with EGF.
Fig. 8. Effect of TGF/31 on TGFa mRNA Stability
Sparse bovine anterior pituitary-derived cells were cultured
for 4 days. The TGF/31 exposure occurred over the last 36 h
of the 4-day incubation. Cells cultured with or without TGF/31
were treated with actinomycin-D (5.0 (ig/m\) for 0 , 1 , 3, 5, and
8 h, after which RNA was extracted from the cells. Two
micrograms of polyadenylated RNA were analyzed by Northern blot, and the nylon membrane was probed simultaneously
with human TGFa cDNA and hexosaminidase-A cDNA. After
autoradiography, the hybridized probes were eluted from the
blot. The Northern blot was then reprobed with the c-myc
cDNA.
with TGF/31, actinomycin-D (5 Mg/ml) was added to the
cell cultures, and RNA was extracted from the cells at
the indicated times after the addition of actinomycin-D.
No decay in the level of TGFa mRNA was detected
during the first 8 h of actinomycin-D treatment in cells
treated in the absence or presence of TGF/31 (Fig. 8).
The c-myc RNA level was greatly reduced after a 1-h
treatment with actinomycin-D and was absent after a
3-h treatment (Fig. 8). These results indicate that the
TGF01 Down-Regulates TGFa
TGFa mRNA is very stable, and this stability is not
significantly altered by TGF/31 treatment.
DISCUSSION
The autocrine loop model was initially proposed to
explain how tumor cells may gain a growth advantage
over normal cells by autoregulation of their own growth
(33). This model has recently been expanded to include
two additional concepts. First, such autocrine loops
exist in nontransformed cells, and second, such a loop
can also autoregulate growth factor expression. Indeed,
TGFa is one of several growth factors (34) known to
up-regulate its own expression, at least in normal anterior pituitary cells and keratinocytes (18, 23). Since
such a positively acting autoregulatory mechanism
could result in growth factor overexpression, an inhibitory mechanism must be postulated, particularly in light
of recent reports on tumor formation in transgenic mice,
which overexpress TGFa (25-27). The results of this
study suggest that TGF/31 may be such a negative
regulator of TGFa production in the cultured anterior
pituitary-derived cells. That is, TGF/31 treatment of
these cells results in a time- and dose-dependent decrease in TGFa mRNA and the rate of TGFa secretion
into the culture medium. In addition, EGF receptor
mRNA is concurrently decreased upon TGF/31 treatment. The decrease in expression of both the EGF
receptor and its ligand TGFa could have a multiplicative
effect, resulting in a marked decrease in ligand-stimulated signalling through the EGF receptor.
The delayed growth response of fibroblasts to TGF/31
has been postulated to result from the stimulation of
platelet-derived growth factor expression by TGF/31
and autocrine growth stimulation through the plateletderived growth factor receptor (35). Thus, it was tempting to postulate that TGF/31 mediated its growth inhibitory effects on the pituitary cells by a converse mechanism, that is through its negative effect on the expression of TGFa and its receptor. However, the inability of
exogenously added EGF to completely reverse the
growth inhibition by TGF/31 suggests that other elements in the growth stimulatory pathway are involved.
The growth inhibitory effects of TGF/31 in keratinocytes
and Mv1Lu lung epithelial cells have recently been
linked to the retinoblastoma gene product (36, 37).
Furthermore, TGF/31 inhibits c-myc gene expression
and cellular proliferation of keratinocytes (38). Since
blockade of c-myc translation with antisense oligonucleotides prevents keratinocyte proliferation, the downregulation of c-myc by TGF/31 may also be sufficient to
inhibit growth (39). Treatment of the pituitary cells with
TGF/31 inhibits c-myc gene expression, and as in keritinocytes, this c-myc down-regulation might be responsible for the observed growth inhibition. However, in
another cell system, TGF/31 inhibition of a-thrombin or
fibroblast growth factor-induced proliferation of Go-arrested Chinese hamster lung fibroblasts occurs without
1443
inhibition of the induction of c-myc by these mitogens
(40). Thus, a decrease in c-myc gene expression may
not be a generalized phenomenon required for TGF/31
growth inhibition.
The failure of EGF to restore the growth of cells
inhibited by TGF/31 did not result from a global blockade
to EGF responsiveness. TGFa expression, which had
been down-regulated by TGF/31, could be restimulated
by EGF. This observation suggests that the mechanism
by which TGF/31 regulates TGFa mRNA is distinct from
the mechanism of growth inhibition.
While it appears that growth inhibition by TGF/31 does
not result directly from decreased TGFa expression,
the converse may hold; that is, the growth inhibition
may cause the decrease in TGFa expression. However,
the following observations suggest a more specific
mechanism. First, both c-myc and TGFa down-regulation occur before the decrease in cellular proliferation.
Second, the down-regulation of these genes is relatively
specific, in that TGF/31 has no effect on actin or hexosaminidase-A mRNA levels. Third, keratinocytes,
whose growth and c-myc expression are inhibited by
TGF01, fail to decrease TGFa expression in response
to TGF/31 (41), indicating that growth inhibition does
not necessarily result in decreased TGFa expression.
The effect of TGF/31 on smooth muscle cell proliferation was shown to be density dependent (31). Smooth
muscle cells differentially express the various TGF01
receptors as a function of cell density, and the receptor
profile appears to dictate the effect of TGF/31 on these
cells. Qualitatively, the effect of TGF/31 on TGFa mRNA
in the pituitary cells was density independent, although
the effect was more pronounced in sparse cells, it is
not known which type of TGF/31 receptors is expressed
on these pituitary cells or whether the receptor profile
is cell density dependent. Investigations in this area
may provide us with information on which TGF/31 receptor is involved with the down-regulation of TGFa
mRNA.
The TGFa mRNA in pituitary and MDA468 cells (24)
is very stable. The decrease in cellular content of TGFa
mRNA in response to TGF/31 could have resulted from
either a decrease in the rate of TGFa gene transcription
or a decrease in the stability of the mRNA. Our experiments show that TGF/31 does not appear to alter this
marked stability of the TGFa mRNA, suggesting that
the observed decrease results from an attenuation of
the synthesis rate. Since the half-life of this mRNA is
so long, any decrease in synthesis would not be reflected rapidly in the TGFa mRNA level, which might
partially explain the long latency between TGF/31 exposure and TGFa down-regulation. Interestingly, this
argument does not hold for the delay in the c-myc
response to TGF/31, since this mRNA has a half-life less
than 1 h in these cells, yet the decrease in c-myc mRNA
was not evident until TGF/31 exposure had occurred for
24 h. This latency of the c-myc response implies, as
previously suggested (39), the presence of an intervening step(s) before the attenuation of c-myc gene transcription.
MOL ENDO-1991
1444
The mechanisms by which TGFa gene transcription
is regulated have not yet been delineated. We (42) and
others (43, 44) have cloned the 5'-flanking region of
the TGFa gene and noted a remarkable lack of known
response elements that might readily account for the
ability of this gene to respond transcriptionally to estrogen (45, 46), phorbol ester (23, 24, 47), and EGF.
Indeed, our laboratory has shown that a 313-base pair
(bp) proximal segment of the gene, lacking AP-2 sites
but containing five SP1 sites, remains EGF and TPA
responsive. A more distal segment of the rat (44) and
human (42) gene markedly attenuates basal and EGFstimulated TGFa: transcription; however, the TGF/31responsive segment has not been defined. The TGFa
5'-flanking sequence thus far examined (1100 bp) does
not contain an element similar to the recently defined
10-bp TGF/31 -negative response element in the transin/
stromelysin gene (48). TGF/31 blocks the ability of the
transin gene to respond positively to EGF, whereas the
TGFa gene remains responsive to EGF after TGF/31
down-regulation. These differences between the TGFa
and transin genes suggest that TGF/31 regulates these
genes through different elements. The availability of this
pituitary cell model and the TGFa gene will make it
possible to define the TGF/31 response element in the
TGFa gene and further define the mechanisms regulating its expression.
MATERIALS AND METHODS
Materials
Culture media, serum, and trypsin were obtained from Gibco
(Grand Island, NY). TGF/31, was purchased from R&D Systems (Minneapolis, MN). [32P]dATP was obtained from Amersham (Arlington Heights, IL). Collagen was obtained from Collagen Corp. (Palo Alto, CA). Actinomycin-D was purchased
from Pharmacia (Uppsala, Sweden). Mouse EGF was purified
from male Swiss Webster mouse submaxillary glands (49).
Pituitary Cell Cultures
Primary cultures of anterior pituitary cells were prepared from
glands of freshly slaughtered calves, as previously described
(50). After an initial overnight plating in 2% calf serum-Dulbecco's Modified Eagle's Medium (DMEM), the cells were grown
in a serum-free medium consisting of DMEM-F-12 (1:1), 10
mg/liter transferrin, 10 mg/liter insulin, 10 mM HEPES, 50 ITIM
sodium selenite, 10 mg/liter ascorbic acid, 100 mg/liter penicillin, 10 mg/liter gentamicin, and 2.5 mg/liter amphotericin.
This medium, except for the amphotericin, was used for all
subsequent cell maintenance. Once the primary cells were
confluent, they were subcultured at a split ratio of 1:5 into 15cm tissue culture plates, maintained in serum-free medium,
and grown to confluence. At this stage, cells were frozen for
subsequent use. Secondary cultures were established by
thawing the primary cells and subculturing the resultant confluent plate. Secondary cells were grown on collagen-coated
plates.
Cellular Proliferation Studies
Cells were subcultured at a density of 5.0 x 104 cells/35-mm
culture dish in 2% calf serum-DMEM. After an overnight incu-
Vol5No. 10
bation, the medium was replaced with serum-free medium
alone, 3.0 ng/ml TGF/31, or 3.0 ng/ml TGF/31 and 10~9 M EGF
together. Cells were treated with fresh TGF/31 and EGF daily.
Cells were counted using a Coulter counter (Hialeah, FL).
Preparation and Analysis of RNA
Total RNA was isolated from the cell cultures by acid guanidinium thiocyanate-phenol-chloroform extraction (51), and polyadenylated RNA was prepared by oligo(dT)-cellulose chromatography (52). Poly(A)+ RNA (2 ^g/lane) was size-fractionated through a 1 % agarose-6% formaldehyde gel, ethidium
bromide stained to assess the integrity and migration of the
RNA, then transblotted onto a GeneScreen nylon membrane.
The RNA was cross-linked to the membrane by UV irradiation
for 3 min with a germicidal General Electric (Toronto, Ontario,
Canada) G15T8 15 W tube at a distance of 12 cm (53). The
membrane was prehybridized for 6 h, then probed for 16 h at
42 C with the appropriate labeled cDNA. Complementary DNA
probes were labeled by the random hexamer primer method
to a specific activity of approximately 2.0 x 109 cpm/mg (54).
The TGFa probe consisted of a 900-bp sequence from the
open reading frame of human TGF« (14), the EGF receptor
probe consisted of a 760-bp sequence corresponding to the
internal domain of the human EGF receptor (55), the c-myc
probe consisted of a 1.4-kilobase (kb) fragment containing the
3' exon of human c-myc (56), the hexosaminidase-A probe
consisted of an approximately 1.6-kb sequence of the asubunit of hexosaminidase-A, a lysosomal enzyme (57), and
the /3-actin probe consisted of a 2-kb fragment from the
chicken /3-actin cDNA (58). After hybridization, the membranes
were washed at high stringency before autoradiography using
Kodak X-Omat AR film (Eastman Kodak, Rochester, NY) with
a DuPont Cronex intensifying screen (Wilmington, DE) at
- 7 0 C.
Elution of Hybridized Probes
Hybridized probes were routinely eluted from the Northern
blots by boiling the membrane for 4 min. After elution, the blot
was prehybridized and hybridized as described above.
Acknowledgments
Received April 3, 1991. Revision received June 13, 1991.
Accepted July 3,1991.
Address requests for reprints to: Jeffrey E. Kudlow, Department of Medicine, Division of Endocrinology and Metabolism, University of Alabama, UAB Station, Birmingham, Alabama 35294.
This work was supported by a grant (to J.E.K.) and a
graduate student award (to S.G.M.) from the Medical Research
Council of Canada.
REFERENCES
1. Derynck R 1986 Transforming growth factor-a: structure
and biological activities. J Cell Biol 32:293-304
2. Marquardt H, Hunkapillar MW, Hood LE, Todaro GJ 1984
Rat transforming growth factor type 1: structure and
relation to epidermal growth factor. Science 223:10791082
3. Carpenter G, Stoscheck CM, Preston YA, DeLarco JE
1983 Antibodies to the epidermal growth factor receptor
block the biological activities of sarcoma growth factor.
Proc Natl Acad Sci USA 80:5627-5630
4. DeLarco JE, Todaro GJ 1978 Growth factors from murine
TGF/31 Down-Regulates TGFa
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
sarcoma virus-transformed cells. Proc Natl Acad Sci USA
75:4001-4005.13.1
Marquardt H, Todaro GJ 1982 Human transforming
growth factors: production by a melanoma cell line, purification and initial characterization. J Biol Chem
257:5220-5225
Massague J 1983 Epidermal growth factor-like transforming growth factor 1. Isolation, chemical characterization
and potentiation by other transforming factors from feline
sarcoma virus-transformed rat cells. J Biol Chem
258:13606-13613
Massague J 1983 Epidermal growth factor-like transforming growth factor II. Interaction with epidermal growth
factor receptors in the human placenta membranes and
A431 cells. J Biol Chem 258:13614-13620
Samsoondar J, Kobrin MS, Kudlow JE1986 a-Transforming growth factor secreted by untransformed bovine anterior pituitary cells in culture. I. Purification from conditioned medium. J Biol Chem 216:14408-14413
Matrisian LM, Pathak M, Magun BE 1982 Identification of
an epidermal growth factor-related transforming growth
factor from rat fetuses. Biochem Biophys Res Commun
107:761-769
Stromberg K, Pigott DA, Ranchalis JE, Twardzik DR 1982
Human term placenta contains transforming growth factors. Biochem Biophys Res Commun 106:354-361
Twardzik DR, Ranchalis JE, Todaro GJ 1982 Mouse
embyonic transforming growth factors related to those
isolated from tumor cells. Cancer Res 42:590-593
Goustin AS, Leof EB, Shipley GD, Moses HL1986 Growth
factors in cancer. Cancer Res 46:1015-1029
Kobrin MS, Asa SL, Samsoondar J, Kudlow JE 1987
Alpha-transforming growth factor in bovine anterior pituitary gland: secretion by dispersed cells and immunohistochemical localization. Endocrinology 121:1412-1416
Kudlow JE, Leung AWC, Kobrin MS, Paterson AJ, Asa
SL 1989 Transforming growth factor-a in the mammalian
brain: immunohistochemical detection in neurons and
characterization of its mRNA. J Biol Chem 264:38803883
Lee DC, Rose TM, Webb NR, Todaro GJ 1985 Cloning
and sequence analysis of a cDNA for rat transforming
growth factor-a. Nature 313:489-491
Wilcox JN, Derynck R 1988 Localization of cells synthesizing TGFa mRNA in the mouse brain. J Neurosci
8:1910-1904
Kudlow JE, Kobrin MS, Purchio AF, Twardzik DR, Hernandez FR, Asa SL, Adashi EY1987 Ovarian transforming
growth factor-a gene expression: immunohistochemical
localization to the theca-interstitial cells. Endocrinology
121:1577-1579
Coffey Jr RJ, Derynck R, Wilcox JN, Bringman TS, Goustin AS, Moses HL, Pittelkow MR 1987 Production and
auto-induction of transforming growth factor-a in human
keratinocytes. Nature 328:817-820
Madtes DK, Raines EW, Sakariassen KS, Assoian RK,
Sporn MB, Bell Gl, Ross R 1988 Induction of transforming
growth factor-a in activated human alveolar marcrophages. Cell 53:285-293
Rappolee DA, Mark D, Banda MJ, Werb Z 1988 Wound
marcrophages express TGFa and other growth factors in
vivo: analysis by mRNA phenotyping. Science 241:708712
Mueller SG, Paterson AJ, Kudlow JE 1990 Transforming
growth factor a in arterioles: cell surface processing of its
precursor by elastases. Mol Cell Biol 10:4596-4602
Kobrin MS, Samsoondar J, Kudlow JE 1986 a-Transforming growth factor secreted by untransformed bovine anterior pituitary cells in culture. II. Identification using a
sequence-specific monoclonal antibody. J Biol Chem
261:14414-14419
Mueller SG, Kobrin MS, Paterson AJ, Kudlow JE 1989
Transforming growth factor-a expression in the anterior
pituitary gland: regulation by epidermal growth factor and
1445
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
phorbol ester in dispersed cells. Mol Endocrinol 3:976983
Bjorge JD, Paterson AJ, Kudlow JE 1989 Phorbol ester
or epidermal growth factor (EGF) concurrently stimulate
the accumulation of the mRNA for the EGF receptor and
its ligand transforming growth factor-a in a breast cancer
cell line. J Biol Chem 264:4021-4027
Jhappan C, Stahle C, Harkins RN, Fausto N, Smith GH,
Merlino GT1990 TGFa overexpression in transgenic mice
induces liver neoplasia and abnormal development of the
mammary gland and pancreas. Cell 61:1137-1146
Matsui Y, Halter SA, Holt JT, Hogan BLM, Coffey RJ
1990 Development of mammary hyperplasia and neoplasia in MMTV-TGFa transgenic mice. Cell 61:1147-1155
Sandgren EP, Luetteke NC, Palmiter RD, Brinster RL,
Lee DC 1990 Overexpression of TGFa in transgenic mice:
induction of epithelial hyperplasia, pancreatic metaplasia,
and carcinoma of the breast. Cell 61:1121-1135
Rosenthal A, Lindquist PB, Bringman TS, Goeddel DV,
Derynck R 1986 Expression in rat fibroblasts of a human
transforming growth factor-a cDNA results in transformation. Cell 46:301-309
Bascom CC, Sipes NJ, Coffey RJ, Moses HL 1989 Regulation of epithial cell proliferation by transforming growth
factors. J Cell Biochem 39:25-32
Roberts AB, Sporn MB 1988 Transforming growth factor
beta. Adv Can Res 51:107-145
Goodman LV, Majack RA 1989 Vascular smooth muscle
cells express distinct transforming growth factor-beta receptor phenotypes as a function of cell density in culture.
J Biol Chem 264:5241-5244
Kerr LD, Olashaw NE, Matrisian LM 1988 Transforming
growth factor beta! and cAMP inhibit transcription of the
epidermal growth factor- and oncogene-induced transin
RNA. J Biol Chem 263:16999-17005
Sporn MB, Todaro GJ 1980 Autocrine secretion and
malignant transformation of cells. New Engl J Med
303:878-880
Derynck R 1988 Transforming growth factor a. Cell
54:593-595
Leof EB, Proper JA, Goustin AS, Shipley GD, DiCorleto
PE, Moses HL 1986 Induction of c-sis mRNA and activity
similar to platelet-derived growth factor by transforming
growth factor-beta; a proposed model for indirect mitogenesis involving autocrine activity. Proc Natl Acad Sci
USA 83:2453-2457
Pietenpol JA, Stein RW, Moran E, Yaciuk P, Schlegel R,
Lyons RM, Pittelkow MR, Munger K, Howley PM, Moses
HL 1990 TGF/31 inhibition of c-myc transcription and
growth in keratinocytes is abrogated by viral transforming
proteins with pRB binding domains. Cell 61:777-785
Laiho M, DeCaprio JA, Ludlow JW, Livingston DM, Massague J 1990 Growth inhibition by TGF/31 linked to
suppression of retinoblastoma protein phosphorylation.
Cell 62:175-185
Coffey Jr RJ, Bascom CC, Sipes NJ, Graves-Deal R,
Weissman BE, Moses HL 1988 Selective inhibition of
growth-related gene expression in murine keratinocytes
by transforming growth factor /3. Mol Cell Biol 8:30883093
Pietenpol JA, Holt JT, Stein RW, Moses HL 1990 Transforming growth factor betai suppression of c-myc gene
transcription: role in inhibition of keratinocyte proliferation.
Proc Natl Acad Sci USA 87:3758-3762
Chambard J-C, Pouyssegur J 1988 TGF-/3 inhibits growth
factor-induced DNA synthesis in hamster fibroblasts without affecting the early mitogenic events. J Cell Physiol
135:101-107
Cook PW, Coffey Jr RJ, Magun BE, Pittelkow MR, Shipley
GD 1990 Expression and regulation of mRNA coding for
acidic and basic fibroblast growth factor and transforming
growth factor a in cells derived from human skin. Mol
Endocrinol 4:1377-1385
Raja RH, Paterson AJ, Shin TH, Kudlow JE 1991 Tran-
Vol5No. 10
MOL ENDO-1991
1446
43.
44.
45.
46.
47.
48.
49.
50.
scriptional regulation of the human transforming growth
factor-a gene. Mol Endocrinol 5:514-520
Jakobovits EB, Schlokat U, Vannice JL, Derynck R, Levinson AD 1988 The human transforming growth factor
alpha promoter directs transcription initiation from a single
site in the absence of a TATA sequence. Mol Cell Biol
8:5549-5554
Blasband AJ, Rogers KT, Chen X, Azizkhan JC, Lee DC
1990 Characterization of the rat transforming growth
factor alpha gene and identification of promoter sequences. Mol Cell Biol 10:2111-2121
Dickson RB, Bates SE, McManaway ME, Lippman ME
1986 Characterization of estrogen responsive transforming activity in human breast cancer cell lines. Cancer Res
46:1707-1713
Liu SC, Sanfilippo B, Perroteau I, Derynck R, Salomon
DS, Kidwell WR 1987 Expression of transforming growth
factor-a (TGFa) in differentiated rat mammary tumors:
estrogen induction of TGFa production. Mol Endocrinol
1:683-692
Pitteikow MR, Lindquist PB, Abraham RT, Graves-Deal
R, Derynck R, Coffey Jr RJ1989 Induction of transforming
growth factor-a expression in human keratinocytes by
phorbol esters. J Biol Chem 264:5164-5171
Kerr LD, Miller DB, Matrisian LM1990 TGF/31 inhibition of
transin/stromelysin gene expression is mediated through
a Fos binding sequence. Cell 61:267-278
Savage Jr CR, Cohen S 1972 Epidermal growth factor
and a new derivative. Rapid isolation procedures and
biological and chemical characterization. J Biol Chem
247:7609-7611
Kudlow JE, Gerrie BM 1983 Production of growth factor
51.
52.
53.
54.
55.
56.
57.
58.
activity by cultured bovine anterior pituitary cells. Endocrinology 113:104-110
Chomczynski P, Sacchi N 1987 Single-step method of
RNA isolation by acid guanidinium thiocynate-phenol-chloroform extraction. Anal Biochem 162:156-159
Aviv H, Leder P 1972 Purification of biolgically active
globin messenger RNA by chromatography on oligothymidylic acid-cellulose. Proc Natl Acad Sci USA 69:14081412
Khandjian EW 1986 UV crosslinking of RNA to nylon
membrane enhances hybridization signals. Mol Biol Rep
11:107-115
Feinberg AP, Vogelstein B 1983 A technique for radiolabeling DNA restriction endonuclease fragments to high
specific activity. Anal Biochem 132:6-13
Lin CR, Chen WS, Kruiger W, Stolarsky LS, Weber W,
Evans RM, Verma IM, Gill GN, Rosenfeld MG 1984
Expression cloning of human EGF receptor complementary DNA: gene amplification and three related messenger
RNA products in A431 cells. Science 224:843-848
Favera RD, Wong-Staal F, Gallo RC 1982 one gene
amplification in promyelocytic leukaemia cell line HL-60
and primary leukaemic cells of the same patient. Nature
299:61-63
Korneluk RG, Mahuran DJ, Neote J, Klavins MH, O'Dowd
BF, Tropak M, Willard HF, Anderson MJ, Lowden JA,
Gravel RA 1986 Isolation of cDNA clones coding for the
alpha-subunit of human beta-hexosaminidase. Extensive
homology between the alpha- and beta-subunits and studies on Tay-Sachs disease. J Biol Chem 261:8407-8413
Cleveland DW, Lopata MA, MacDonald RJ, Cowan NJ,
Rutter WJ, Kirshner MW 1980 Number and evolutionary
conservation of alpha- and beta-tubulin and cytoplasmic
beta- and gamma-actin genes using specific cloned cDNA
probes. Cell 20:95-105